# Flow Prd

> Creation of a PRD (Product Requirements Document) with step-by-step discipline, single source of truth for scope.

- **Type:** Skill
- **Install:** `agentstack add skill-edouard-claude-pi-flow-skills-flow-prd`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [edouard-claude](https://agentstack.voostack.com/s/edouard-claude)
- **Installs:** 0
- **Category:** [AI & ML](https://agentstack.voostack.com/c/ai-and-ml)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [edouard-claude](https://github.com/edouard-claude)
- **Source:** https://github.com/edouard-claude/pi-flow-skills/tree/main/skills/flow/flow-prd

## Install

```sh
agentstack add skill-edouard-claude-pi-flow-skills-flow-prd
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# flow-prd — PRD single source of truth

You are a peer PM facilitator. You produce an **append-only** document step by step, with a validation menu at each step. No skipping. The user validates each section before the next.

## When to use

- After `flow-brief` (greenfield) or `flow-introspect` (brownfield)
- Before `flow-architecture` — the PRD feeds the architecture

## Inputs (read BEFORE asking questions)

- `.agents/planning/product-brief.md` if present (greenfield)
- `.agents/planning/current-state.md` if present (brownfield)
- `.agents/project-context.md` if present (technical constraints)
- Existing PRD if update mode

## Process — step-by-step discipline

### Step 1 — Mode & init
"Do you want: (a) PRD from scratch, (b) update of an existing PRD?" → init `.agents/planning/prd.md` with `stepsCompleted: []` frontmatter.

### Step 2 — Executive summary & problem
Questions in batches of 5 max:
- Detailed problem (beyond the brief)
- Business context / external constraints
- Why now
- Status quo and cost of inaction

**Append to the PRD, update `stepsCompleted`, present menu**:
```
1. Continue (Step 3 — Users & user stories)
2. Revise the current section
3. Pause
```

### Step 3 — Users & user stories
- Personas (primary, secondary)
- Priority user stories (format "As a X, I want Y, so that Z")
- User journey if relevant

**Append → validation menu.**

### Step 4 — Success metrics
- Measurable KPIs (north star + supporting)
- Definition of "success" for v1
- Business metrics + product metrics

**Append → validation menu.**

### Step 5 — Scope
- Scope IN (v1 features)
- Scope OUT (explicit — what we will NOT do)
- Phasing if multiple releases planned

**Append → validation menu.**

### Step 6 — Constraints & dependencies
- Technical constraints (perf, security, integrations)
- Business constraints (budget, time, compliance)
- External dependencies (third-party APIs, teams, upstream deliverables)

**Append → validation menu.**

### Step 7 — Non-goals & risks
- Non-goals (areas explicitly out of scope)
- Identified risks + planned mitigations

**Append → validation menu.**

### Step 8 — Finalize
- Global proofread
- Ask for explicit validation: "Is the PRD ready for `flow-architecture`?"
- Optionally offer an LLM-optimized distillate (1 page)

## Output

`.agents/planning/prd.md` with tracking frontmatter:
```yaml
---
status: draft | ready
stepsCompleted: [1, 2, 3, ...]
lastUpdated: 2026-MM-DD
---
```

Sections: Executive Summary, Problem, Users + Stories, Success Metrics, Scope, Constraints, Non-goals, Risks.

## Next

Explicit validation required before `/flow-architecture`. If update mode, also suggest `/flow-course-correct` if it impacts an in-progress sprint.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [edouard-claude](https://github.com/edouard-claude)
- **Source:** [edouard-claude/pi-flow-skills](https://github.com/edouard-claude/pi-flow-skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-edouard-claude-pi-flow-skills-flow-prd
- Seller: https://agentstack.voostack.com/s/edouard-claude
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.